Cut Herringbone Waste to Under 8 Percent with a Rubi DU-200 Bridge Saw on Chevron Layouts

Chevron waste starts at the two 45-degree cuts on each picket. Each end cut removes a triangular sliver, and the size of that sliver grows with picket width. When a 600 mm porcelain plank is cut into 90 mm chevron pieces by hand, the sliver loss combines with chipped-edge rejects, leaving many crews in the 14 to 18 percent waste range.

The Rubi DU-200 changes the yield because the sliding bridge carries the head through the same line of travel and holds a fixed 45-degree miter fence. Pickets come off the saw with the same angle and length, so chipped mitres fall sharply and the offcut triangles can be planned into the next cuts. The job gains usable pieces through repeatability, clean edges, and a nest that can be followed across the slab.

A fence and length stop that hold their setting are what let the offcut triangles be paired against one another on the next slab. Keep the batch tight and consistent, and the sub-8 percent figure follows from that consistency more than from any single clever cut.

Where chevron differs from classic herringbone

Classic herringbone is made from square-ended rectangles laid at 90 degrees to each other, so the pattern can be cut with very little geometric waste. Chevron looks similar once installed, yet the mitred ends change the cutting problem, which is why many estimators still pad orders to 15 percent when the work will be done on a manual cutter.

The DU-200 uses a water-cooled continuous-rim blade beneath a sliding head, and the head runs on the same rail every pass. Once the 45-degree stop is set and a length gauge is clamped to the frame, the saw produces mirrored left and right pickets without measuring each piece again.

Cut the left-hand pickets from the top of the slab, then cut the right-hand pickets from the triangular remains created by the first sequence. Two offcuts that would usually go into a bucket can combine into one full picket, and that pairing is where the yield comes from.

On a 1200 by 600 porcelain slab cut into 90 mm chevron, careful nesting on the DU-200 can produce 11 to 12 usable pickets. A manual scribe setup commonly yields 8 or 9. That difference accounts for the movement from typical chevron padding toward the 8 percent waste range.

Blade choice and kerf drift

A cheap segmented blade chips the glazed edge of porcelain, and each chipped mitre becomes a reject that raises the waste percentage before the layout geometry is finished. A Rubi CPC blade, or a comparable continuous-rim wet blade around 200 to 250 mm, keeps the cut cool and leaves an edge clean enough for a 2 mm grout joint. Chevron exposes its mitre seams under raking light, so a damaged 45-degree corner can read across the whole floor.

Blade wear affects the cut as the batch grows. Exact kerf change varies with blade quality, feed pressure, water flow, porcelain hardness, and operator pace, yet site experience on 10 mm porcelain often shows measurable drift after a few hundred linear metres. A worn blade can widen the kerf by a couple of tenths of a millimetre, a small amount on one cut that becomes important across 200 mitred ends.

Dress or replace the blade before starting a chevron batch if the blade is already near that point in its life. Logging the metres cut on each blade gives the crew a record to compare against the nesting tolerance before the grout stage exposes the issue.

The water feed on the DU-200 earns its keep. If the tray runs low or the pump draws air, the rim can glaze over, the cut slows, and the leading corner of the porcelain is more likely to fracture. Keep the tray topped and the pump submerged throughout the batch.

A 30 square metre floor in slabs and money

Take a 30 m2 living space, 90 mm chevron, and 1200 by 600 slabs at 10 mm thickness. Each slab covers 0.72 m2 before cutting. With the usual 15 percent padding, the order is 30 divided by 0.72 times 1.15, rounded to 48 slabs.

At 8 percent waste on the DU-200, the calculation becomes 30 divided by 0.72 times 1.08, which rounds to 45 slabs. That saves three slabs on a 48-slab order. At 40 per slab, the recovered material value is 120; at 90 per slab, it is 270. The saw is a fixed cost that can be spread over later jobs, so the material recovery continues after the first floor.

Labour adds another saving. Manual mitring and dressing a chevron picket takes a competent tiler about 90 seconds or more per piece, depending on tile hardness, blade condition, and the amount of chipping. With a bridge saw and a length stop, the same cut can be made in under 25 seconds, again with variation by operator and material. Across more than 400 pickets on a 30 m2 floor, the time difference becomes several working hours.

Nesting the offcuts before the first cut

Crews often cut all the full pickets first and then try to rescue the triangles afterward. By that stage, the useful geometry may already be gone.

Plan the nest before the first cut. Draw the slab, mark the left-hand picket runs, and then mark where the mirrored right-hand pickets will sit inside the residual triangles. The cutting sequence needs to follow that drawing, because the value is held in the order of operations as much as in the saw itself.

Mark the two hands as they leave the saw. A wax pencil is enough, provided the mark is visible and consistent. Left-hand and right-hand pickets can look deceptively interchangeable when they are stacked wet beside the saw.

A flipped chevron picket reverses the mitre direction and interrupts the zig-zag. The error may be hard to see while adhesive is fresh, especially when the piece is surrounded by wet porcelain and spacer clips. Once the adhesive has grabbed, the repair costs far more than the pencil mark would have.

The same discipline applies to the triangle pile. Keep offcuts from the left-hand sequence together and feed them into the right-hand sequence while their reference edges are still known. Randomised offcuts usually lose the clean pairing that made the nest possible.

On larger floors, check the nest after the first few slabs. If the first batch is producing fewer pieces than the drawing predicted, the problem is usually fence movement, slab variation, kerf growth, or a cut sequence that has drifted from the plan.

Bedding the pattern so the seams stay true

Chevron is unforgiving on a moving substrate because the mitred seams run continuously across the floor. Deflection tends to show at the joint. On a timber subfloor or a fresh underfloor heating screed, an uncoupling layer such as the Schluter Ditra system or a comparable Rubi uncoupling membrane absorbs differential movement between substrate and tile. The membrane also allows the screed to finish curing without transferring shrinkage stress into the porcelain.

Use a large-format tile adhesive, C2 class or better, with a notched trowel sized to the slab, then back-butter each picket so a 1200 mm plank reaches full contact. Voids beneath a chevron picket are where a seam can eventually lip. On an underfloor heating screed, the screed should reach its stated moisture content before tiling, because drying under the porcelain can pull the mitres apart at the seam.

Grout spacing changes both the look and the tolerance. A 2 mm joint gives the floor a near-seamless zig-zag and flatters the mitred line, yet it leaves little room for lippage, so cutting and bedding must be tight. A 3 mm joint forgives more variation and softens the crispness of the pattern. Where the floor meets a wet room waterproofing membrane at a shower threshold, plan the picket on the transition so a full mitre, rather than a thin sliver, sits over the membrane lap.

The tolerance stack left by the slab batch

Fence drift, blade kerf, and slab dimensional variance add together across the run. Porcelain slabs from the same batch can vary by tenths of a millimetre, with larger differences possible across batches, so measure the actual slab before setting the length stop and check it again when a new pallet is opened.

When the second pallet arrives, redraw the nest against a slab pulled from that stack before committing the length stop, rather than trusting the drawing made from the first delivery. A nest that fit cleanly on paper can start crowding the mirror cuts the moment the material underneath it shifts by a fraction, and catching that at the drawing board is cheaper than catching it at the grout line.